A Conversation with Dave Kranzler – Co-Author of New Industry Reference Book
Our SVP of product development, Dave Kranzler, contributed to key chapters on AMI and Smart Grid IoT in an important new book on the changing energy industry, Smart Grids: Advanced Technologies and Solutions. We had a conversation with him recently on the promise and challenges of communications in the smart grid era.
Q: How did you become involved with the second edition of the book?
Kranzler: The book is edited by Stuart Borlase and a mutual friend told me that he was looking for help on editing a few chapters, so I volunteered. I have a general interest in IoT and deal with it in my work every day, so I thought it’d be a good way to engage with some industry experts and help educate others on these topics through a published book.
Q: One of the chapters you co-authored was the one on Internet of Things. Why did the book’s author feel it was important to include this topic in the book?
Kranzler: We define IoT as a high-volume connectivity of devices, rather than people. Our view is that smart meters and smart grids are really the first volume markets for the IoT. So, no book on smart grids would be complete without a discussion of the technologies and developments that are enabling IoT.
Now, one of the bigger questions you have to ask about any technology is, what problem do you want to solve with something like IoT? The energy industry has already defined several important problems to solve with this technology. One is smart metering and the ability to digitally manage people’s energy usage. Another is the smart grid and the capabilities it provides to improve the efficiency of utilities’ operations across the distribution network.
Consequently, if you look at the installed base of smart metering and smart grid networks out there, it really is the first volume market for IoT.
Q: What do you feel are the biggest challenges surrounding AMI and smart grid IoT today that people should better understand?
Kranzler: Well, I’d say that initial adoption is not one of the challenges. The first generation of advanced metering infrastructure is already in place. We have fairly high smart metering penetration in the U.S. For example, it is estimated somewhere around fifty percent or so, give or take ten.
However, those networks that have been deployed are limited in their capabilities, in terms of what problems they can solve beyond just collecting billing reads. So, we are now entering sort of a second wave of technology delivery and deployment in smart metering, which provides higher bandwidth, lower latency and more ubiquitous connectivity. It solves some of the problems that weren’t fixed during the first-generation AMI and creates the opportunity to run other applications besides just reading billing data from the smart metering network.
Consequently, with these new capabilities now becoming available, we’d like to see adoption really take off, and it hasn’t yet. There are regulatory challenges around offering the right incentives, such that smart metering achieves a much higher level of penetration than it does. There needs to be a better and more natural economic incentive, so that this technology gets proliferated even more broadly than it is today.
Q: Can you talk about some of the changes you’ve seen in smart metering communications during that time? Which trends or changes are most significant, in your view?
Kranzler: The first was the transition from automated meter reading, or AMR, to first generation advanced metering infrastructure, or AMI. It’s moving from essentially a one-way communication network to a two-way communication network. A one-way communication network allows you to read meters, but do very little else.
Forty years ago, my father’s company built automated meter reading products using telephone lines in New Jersey. Trilliant was founded thirty years ago and was doing the same thing in Quebec. That is the baseline from which Trilliant started. Ten years ago, AMI pioneers started deploying networks that had two-way communication, allowing you to push command and control down to the meters , something that you couldn’t do with AMR.
Two-way communication requires smaller, cheaper, higher performance processors you could embed in a meter or device for more processing power. Higher performing two-way networks also require communications protocols that are robust enough to cover a large geographical area. It’s one thing is to send an occasional signal over a phone line; however, it is a much different technical problem to send large amounts of data, at higher speeds and lower latency, across the ether – not just across a neighborhood, but over a city, state or even an entire country.
Now, those four elements: the smaller, higher performance radio; the smaller, cheaper higher performance processor; the communication protocols that operate over a wide geographical area; and the sensor, which in this case is the smart meter itself, are precisely what conspire to make next generation smart metering and smart grid possible. The same tools are required for smart city and IoT, but with different sensors to solve different problems.
Q: In this increasingly interconnected world, how can utilities manage all the different types of communications that they may need?
Kranzler: We stress to the utility world that one size does not fit all; there is no one communications technology that can solve all problems. So, why is that? Because we now have the desire and the ability to do more and solve more problems in the energy network than just reading energy bills.
For example, if my goal is to bill customers, I can read their meter, in theory, only once every thirty days. But, to enable a customer to pre-pay on the meter for their energy, you need the ability to operate almost in real-time.
If a customer adds credit to their account, we must deliver that message to the meter within forty-five minutes. That may not sound tricky if you’re only talking about one meter, however, if there are four million meters, in which ten percent are pre-pay, that is a much bigger problem. That may require different communications technology than the one you use to read that meter every thirty days.
Here’s another example. We talked about driving the penetration of smart metering up from where it is today to everyone having a smart meter. How does the utility do that? If you are the utility and you’ve been told to deploy smart metering services to everyone, do you deploy everyone’s houses at once? Or deploy at the rate of customers who want to have a smart meter? Maybe some customers do not want one. Sometimes meters are inside the house, not outside, so you have to knock on a door and make an appointment.
The utility’s deployment model dictates the requirements of the technology. In the U.S., we have the ability to go block by block and walk into a person’s backyard and replace the meter without them ever knowing. Our RF mesh technologies are great for that and many of the first-generation AMI networks are built on arc mesh.
With a sparse deployment, such as a village, remote area, business park or a situation where you need to make an appointment with the customer, meshes do not work well because they rely on everyone in the area being connected. In this case, long-range radios and cellular technologies are better adapted to the problem than mesh, because they can support a deployment model in which only a few people in an area are connected at a time.
One more: smart grid applications require different communications than smart metering. There are numerous opportunities to improve safety, reliability, uptime and the operating costs under the distribution network, but you must be able to talk to devices scattered over large geographical areas. That is a different type of communications technology used to talk to a large number of meters in a dense, urban area or in a town or village. In addition, distribution automation applications may require a lot more data to be transferred, or require much faster response times, than smart metering does.
Q: How can a utility know what technology to invest in when technology and regulations are always in flux? What priorities should they focus on?
Kranzler: It is critical that utilities look down the road for a roadmap of problems they want to solve before investing in a network. When a regulator commands or incentivizes a utility to solve a problem, it is way too tempting for utilities to purchase a network that only solves that problem. This is risky, because what does the utility do if that network cannot solve the next problem, and the one after that? It is very difficult to go back to the regulator and ask for another large capital investment. Furthermore, many utilities are compensated on return on assets. The more benefit the utilities can receive from a capital investment –the more applications they can run on their network – the more profit they can make.
In the book, one of the key messages is that if a utility thinks about a smart metering network as just a smart metering network, they are more likely to make short-term choices that they’ll regret. However, if utilities think of a smart metering network as a smart grid IoT, that expands their view substantially. Now, the utilities can consider other things they could do with their network regarding problem solving and reducing costs. Therefore, the utilities are more likely to make better choices and earn a much higher return on assets overtime.
Q: What role have communications played historically? What do you believe is in store for the future of smart grid communications?
Kranzler: Historically, communications have been used to lower operational costs or increase operational efficiency. Going forward, the communications network could be used to help the utility better engage with consumers, and better serve the communities in which they operate. Here are just two examples:
Communications networks are a great way to detect theft. In several emerging economies, theft, or what we call non-technical loss, can reach astronomically high portions, around thirty to forty percent of what utilities would expect revenue to be. Thus, the utility cares about preventing theft and they need the technology to detect where the theft is occurring in order to take counter measures. Smart communications networks can help utilities reduce the burden of energy theft, which otherwise would be distributed amongst the ratepayers.
Through decarbonization, you can reduce the overall amount of energy usage or at least reduce it at times when the energy is more expensive or more polluting to produce. This is a current use, and I believe we will see more of it in the future as more of these tools get rolled out. The tools will allow utilities to shape peaks of demand and deal with the duck curve as the sun goes down. These tools will also allow the utilities to integrate renewables more seamlessly into the rest of their grid. Since renewables create a lot of noise, communications networks can help you detect and take counter measures to ensure the grid remains stable.
Q: The book features a chapter on the implementation and operation of an integrated smart grid, and it features a lot of examination of the enabling technologies. Do you see communications as an enabling technology? If so, in what ways?
Kranzler: The four things you need for IoT are: communications, processing, protocols and sensors. If these four things are needed for IoT, then they are also needed for a smart, digital real-time energy grid. If you want to gather real-time data and take and exercise real-time control, you must have real-time communications. Processing power gets pushed closer to the edge of the network to allow you to make decisions and act close to where the problem is. Sensors enable the edge of the network. An IoT is all about placing sensors in more places, so you can gather more data on what’s going on in the physical world. Across the distribution network, an example of adding sensors is measuring voltage at different parts of the network to allow you to optimize how much power and voltage you put out to the network.
Q: Lots of people talk about IoT. What is your definition? What’s the difference between saying “IIoT” and “IoT” – is there a distinction that matters to the electric industry (and utilities in particular)?
Kranzler: I don’t know if my definition of IoT differs from anybody else’s. In its most general sense, an IoT is a global communications network for things to interact, and ideally to interact to solve problems in a useful way, as opposed to interact in negative ways. The Internet can certainly be abused for nefarious purposes; the IoT can be as well.
The IoT is an extremely broad topic. The estimate that Gartner said is that there will be twenty billion devices connected by 2020 and trillions of dollars of business – that encompasses the huge number of target markets.
IIoT, a segment of IoT, is targeted at making industry more efficient. In this context, “industry” includes energy, manufacturing, commercial building management, etc. All of these applications kind of fit into IIoT. The distinction between IIoT and IoT can also be technical, where the technology one would use in smart metering is different from remote patient health management.
IIoT can probably be grouped into a category that includes outdoor applications, things that cover cities, states and countries, just like smart metering and smart grid. That implies certain technology choices. There are some indoor IIoT applications, like inside a manufacturing facility or commercial building, that is better managed to reduce its energy usage, make its occupants more comfortable, or improve its security.
Q: What advancement do you foresee in IoT communications in the near future, within the next five to ten years? And how should utilities anticipate and prepare for those changes?
Kranzler: There are going to be advances in the four factors I previously described: communications, processors, protocols and sensors, but they will be evolutionary in nature. They’ll all get smaller, cheaper, better and faster, but nothing that is revolutionary. I think the major technological change will be making better use of all of the terabytes and petabytes of data that are being collected.
Big data analytics are going to continue to be a large deal, because when you receive a lot of data, it is way more than a human can process. Thus, having machines that provide you real, actionable business insights, out of all of this data, is going to be critical to be able to achieve an ROI out of this network investment.
If you do not deploy data analytics to help you make sense of all of the stuff you’re learning, you are missing out on a large part of the benefit you can get from an IoT. I think that is true across all IoT and particularly true for IIoT.
The second thing would be security enhancements and how the IoT can be hijacked and attacked, so the ability to detect and quarantine security threats is going to be critical to build the ubiquity as well as the trust in IoT. And there, I think the most interesting technologies will be around machine learning and the detection and mitigation of threats. The protocols today are already secure, but we’ve already seen attacks on human Internet and attacks on IoT, where people can still bring a network down, through hacked devices or various other misbehaviors. The ability to detect and shut it down quickly is crucial to IoT taking off.
Those two areas to me are where the more revolutionary change will take place; I think those previous four things in IoT will just be evolutionary.
Q: If readers could only take away one piece of information from this book, what do you believe it should be?
Kranzler: The book is targeted at IIoT. We don’t cover the broader IoT market, so I would say the one thing to take away is that the technology here is being deployed, and it is coming whether you like it or not. Therefore, the wise reader is going to educate themselves about what the current state of the technology is, how it can be used and where it is going. So, they may incorporate it into their business planning. You don’t have the luxury of ignoring it because you’re going to miss out on an opportunity or become uncompetitive.
There are tips, tricks and pitfalls that the reader needs to take advantage of. I wish our customers would all read this book because I think they would have a better appreciation for the issues and problems that need to be solved and opportunities they could be capitalizing upon.
Because quite frankly, and quite selfishly, Trilliant’s best customers are the ones who will understand this the best and who appreciate what we do in our technology to be able to solve these problems. The customers who care only about cost and about solving today’s problems are not our best customers, because they won’t appreciate the added value that we put into our network, and, consequently, make short-term decisions that they will regret.